To determine the resolution and contrast sensitivity of a number of observers under various conditions. To determine the error rates of these observers (radiologists and residents in radiology) when they are shown lesions of known position but of various levels of visibility as determined by the TV 'eye'. To determine the error rates of these observers on lesions of known visibility when the position of the lesion is not known. Thus, if a lesion is known to be visible and it is missed when its position is not known, it was not included in the search pattern. Knowing this the search pattern can be improved.